Project Abstract Overview: Ductal carcinoma in situ (DCIS) comprises approximately 20% of breast cancers diagnosed annually in the US. DCIS is currently treated aggressively with surgery and radiation, or mastectomy. Yet the excellent long-term outcome for many DCIS patients indicates that there is substantial over-treatment of DCIS, and trials of active surveillance are starting. Relevance: The critical question is which cases of DCIS can be safely followed without being removed? One major unmet need is that 1/4 of DCIS cases diagnosed by core needle biopsy actually harbor occult invasive cancers that merit de?nitive treatment. Accurate identi?cation of DCIS cases to exclude these invasive cancers has been identi?ed as one of the major roadblocks to active surveillance. Currently, MRI is viewed as the most promising staging strategy. However, while MRI is twice as sensitive as mammography for breast cancer detection overall, it is not that accurate for selectively detecting invasive cancers in a background of DCIS. Approach: Over two prior funding cycles, we have made key developments in high spatiotemporal resolu- tion breast MRI that are industry standards, with numerous advances in other breast MRI sequences including diffusion-weighted imaging (DWI). Here, we propose to investigate and test two promising areas of MRI technol- ogy to improve discrimination of invasive cancer from DCIS. In Aim 1, we will develop dynamic contrast enhanced MRI with high spatiotemporal resolution sampling, spatiotemporally-sparse compressed-sensing reconstruction, integrated fat-water separation and non-rigid motion correction. We will also incorporate novel pharmacokinetic mapping that includes modeling of arterial bolus dispersion for better lesion discrimination. In Aim 2, we will develop diffusion-weighted MRI with multi-shot methods that use novel shot-to-shot motion correction, fast-spin- echo diffusion imaging for better performance around biopsy markers, and steady-state diffusion for 3D imaging. Finally in Aim 3, we will validate and test the accuracy of these new techniques in large patient studies, to show their bene?t for invasive tumor detection in DCIS patients compared to conventional ACR-standard breast MRI. Summary: The advanced MRI techniques that we propose to develop, validate and test in this proposal have tremendous potential to identify which DCIS patients can safely select active surveillance and which need de?ni- tive treatment. High accuracy for invasive cancer will provide this value with many fewer false-positive biopsies than conventional breast MRI.